Original Articles Open access

Field-based implementation of machine learning forecasting for energy-efficient temperature control in a greenhouse system

Authors

Published
5 June 2026
Section
Original Articles

Abstract

Greenhouse heating systems require control strategies that reduce unnecessary energy use while maintaining crop-safe temperatures under dynamic winter conditions. This study evaluated the practical application of machine learning (ML)-based forecasting for greenhouse temperature control in a single winter greenhouse equipped with a pellet boiler, heat-storage tank, and tube-rail heating system. Data were collected for 55 d under real operating conditions, including internal environmental variables and external weather variables such as external temperature, humidity, dew point, and solar radiation. Prediction and forecasting models were developed using time-series ML algorithms with gap labeling, and iterative learning was conducted using 7 d of training data and 1 d of validation data. The developed models showed an average r² of 0.77 and RMSE of 1.43°C across the evaluated cases. The forecasting results were then applied to a data-driven predictive control (DDPC) strategy to assess whether heating supply could be stopped earlier than in the conventional on/off control scenario. Compared with the baseline control scenario, the DDPC-based strategy showed an estimated 5-15% reduction in heating supply operation time. This value should be interpreted as an operation-time-based estimate under the tested pellet-boiler greenhouse conditions, rather than as a direct measurement of pellet-fuel consumption. The results indicate that 30- and 60-min forecasting horizons can support practical heating-control decisions, whereas longer horizons may increase the risk of low-temperature stress. Therefore, this study emphasizes the field-based implementation of ML forecasting for real-time greenhouse control and clarifies its operational limitations under actual winter conditions.

Supporting Agencies

Korea Institute of Planning and Evaluation for Technology in Food, Agriculture and Forestry (IPET) through the Smart Farm Innovation Technology Development Program funded by Ministry of Agriculture, Food and Rural Affairs (MAFRA) (grant number RS-2025-02221279), National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. RS-2022-NR072396)

Data Availability Statement

The datasets used or analyzed during the current study are available from the corresponding author on reasonable request.

References

  1. Abasilim C, Friedman LS, Martin MC, Madigan D, Perez J, Morera M, et al., 2024. Risk factors associated with indicators of dehydration among migrant farmworkers. Environ Res 251:118633. DOI: https://doi.org/10.1016/j.envres.2024.118633
  2. Aggarwal CC, 2018. Neural networks and deep learning: a textbook. Cham, Springer. DOI: https://doi.org/10.1007/978-3-319-94463-0
  3. Ariga M, Nakayama S, Nishibayasi D, 2018. Machine learning at work. Sebastopol, O’Reilly Media.
  4. Cemek B, Atiş A, Küçüktopçu E, 2017. Evaluation of temperature distribution in different greenhouse models using computational fluid dynamics (CFD). J Agric Sci 32:54-54. DOI: https://doi.org/10.7161/omuanajas.289354
  5. Chang Z, Zhang Y, Chen W, 2019. Electricity price prediction based on hybrid model of adam optimized LSTM neural network and wavelet transform. Energy 187:115804. DOI: https://doi.org/10.1016/j.energy.2019.07.134
  6. Chen J, Xu F, Tan D, Shen Z, Zhang L, Ai Q, 2015. A control method for agricultural greenhouses heating based on computational fluid dynamics and energy prediction model. Appl Energy 141:106-118. DOI: https://doi.org/10.1016/j.apenergy.2014.12.026
  7. Choi W, Lee S, 2023. Performance evaluation of deep learning architectures for load and temperature forecasting under dataset size constraints and seasonality. Energy Build 288:113027. DOI: https://doi.org/10.1016/j.enbuild.2023.113027
  8. Choi YS, Lee HJ, Joung ST, 2012. A design and implementation of web-based green house automation system. J Korea Inst Electron Commun Sci 7:1519-1527.
  9. da Rosa Righi R, Goldschmidt G, Kunst R, Deon C, da Costa CA, 2020. Towards combining data prediction and internet of things to manage milk production on dairy cows. Comput Electron Agric 169:105156. DOI: https://doi.org/10.1016/j.compag.2019.105156
  10. Dahl M, Brun A, Andresen GB, 2017. Using ensemble weather predictions in district heating operation and load forecasting. Appl Energy 193:455-465. DOI: https://doi.org/10.1016/j.apenergy.2017.02.066
  11. Deng Z, Wang X, Jiang Z, Zhou N, Ge H, Dong B, 2023. Evaluation of deploying data-driven predictive controls in buildings on a large scale for greenhouse gas emission reduction. Energy 270:126934. DOI: https://doi.org/10.1016/j.energy.2023.126934
  12. François C, 2017. Deep learning with Python. Shelter Island, Manning Publications.
  13. Frausto HU, Pieters JG, Deltour JM, 2003. Modelling greenhouse temperature by means of auto regressive models. Biosyst Eng 84:147-157. DOI: https://doi.org/10.1016/S1537-5110(02)00239-8
  14. Graves A, 2012. Long short-term memory. In: Supervised sequence labelling with recurrent neural networks. Berlin, Springer pp. 37-45. DOI: https://doi.org/10.1007/978-3-642-24797-2_4
  15. Guzmán CH, Carrera JL, Durán HA, Berumen J, Ortiz AA, Guirette OA, et al., 2019. Implementation of virtual sensors for monitoring temperature in greenhouses using CFD and control. Sensors (Basel) 19:60. DOI: https://doi.org/10.3390/s19010060
  16. He Z, Jiang T, Jiang Y, Luo Q, Chen S, Gong K, et al., 2022. Gated recurrent unit models outperform other machine learning models in prediction of minimum temperature in greenhouse based on local weather data. Comput Electron Agric 202:107416. DOI: https://doi.org/10.1016/j.compag.2022.107416
  17. Hochreiter S, Schmidhuber J, 1997. Long short-term memory. Neural Comput 9:1735-1780. DOI: https://doi.org/10.1162/neco.1997.9.8.1735
  18. Huh M-H, 2017. Representing variables in the latent space. Korean J Appl Stat 30:555-566. DOI: https://doi.org/10.5351/KJAS.2017.30.4.555
  19. Jung D-H, Kim HS, Jhin C, Kim H-J, Park SH, 2020. Time-serial analysis of deep neural network models for prediction of climatic conditions inside a greenhouse. Comput Electron Agric 173:105402. DOI: https://doi.org/10.1016/j.compag.2020.105402
  20. Kadlec P, Gabrys B, 2009. Soft sensors: where are we and what are the current and future challenges? IFAC Proc 42:572-577. DOI: https://doi.org/10.3182/20090921-3-TR-3005.00098
  21. Kim R-W, Hong S-W, Lee I-B, Kwon K-S, 2017. Evaluation of wind pressure acting on multi-span greenhouses using CFD technique, Part 2: Application of the CFD model. Biosyst Eng 164:257-280. DOI: https://doi.org/10.1016/j.biosystemseng.2017.09.011
  22. Kumari P, Toshniwal D, 2021. Long short term memory–convolutional neural network based deep hybrid approach for solar irradiance forecasting. Appl Energy 295:117061. DOI: https://doi.org/10.1016/j.apenergy.2021.117061
  23. Kwon H, Oh KC, Choi Y, Chung YG, Kim J, 2021. Development and application of machine learning‐based prediction model for distillation column. Int J Intell Syst 36:1970-1997. DOI: https://doi.org/10.1002/int.22368
  24. Lederrey G, Lurkin V, Hillel T, Bierlaire M, 2021. Estimation of discrete choice models with hybrid stochastic adaptive batch size algorithms. J Choice Modell 38:100226. DOI: https://doi.org/10.1016/j.jocm.2020.100226
  25. Lee D, Ooka R, Matsuda Y, Ikeda S, Choi W, 2022. Experimental analysis of artificial intelligence-based model predictive control for thermal energy storage under different cooling load conditions. Sustain Cities Soc 79:103700. DOI: https://doi.org/10.1016/j.scs.2022.103700
  26. Liu G, Zhong K, Li H, Chen T, Wang Y, 2024. A state of art review on time series forecasting with machine learning for environmental parameters in agricultural greenhouses. Inf Process Agric 11:143-162. DOI: https://doi.org/10.1016/j.inpa.2022.10.005
  27. López Santos M, Díaz García S, García-Santiago X, Ogando-Martínez A, Echevarría Camarero F, Blázquez Gil G, Carrasco Ortega P, 2023. Deep learning and transfer learning techniques applied to short-term load forecasting of data-poor buildings in local energy communities. Energy Build 292:113164. DOI: https://doi.org/10.1016/j.enbuild.2023.113164
  28. Lv W, Shen C, Li X, 2018. Energy efficiency of an air conditioning system coupled with a pipe-embedded wall and mechanical ventilation. J Build Eng 15:229-235. DOI: https://doi.org/10.1016/j.jobe.2017.11.010
  29. Muller AC, Guido S, 2017. Introduction to machine learning with Python: a guide for data scientists. Sebastopol, O’Reilly Media.
  30. Oh KC, Kim SJ, Park SY, Lee CG, Cho LH, Jeon YK, Kim DH, 2022. Development and verification of smart greenhouse internal temperature prediction model using machine learning algorithm. J Bio-Env Con31:152-162. DOI: https://doi.org/10.12791/KSBEC.2022.31.3.152
  31. Oh KC, Park SY, Kim SJ, Choi YS, Lee CG, Cho LH, Kim DH, 2019. Development and validation of mass reduction model to optimize torrefaction for agricultural byproduct biomass. Renew Energy 139:988-999. DOI: https://doi.org/10.1016/j.renene.2019.02.106
  32. Oh KC, Park SY, Kim SJ, Cho LH, Lee CG, Kim DH, 2024a. Development of a greenhouse environmental forecasting model using machine learning to optimize energy consumption. Available from: https://ssrn.com/abstract=4975488 DOI: https://doi.org/10.2139/ssrn.4975488
  33. Oh KC, Kwon H, Park SY, Kim SJ, Kim J, Kim D, 2024b. Hyperparameter optimization of the machine learning model for distillation processes. Int J Intell Syst 2024:5564380. DOI: https://doi.org/10.1155/2024/5564380
  34. Outanoute M, Lachhab A, Ed-Dahhak A, Selmani A, Guerbaoui M, Bouchikhi B, 2015. A neural network dynamic model for temperature and relative humidity control under greenhouse. Proc. Third Int. Workshop on RFID and Adaptive Wireless Sensor Networks (RAWSN), Agadir; pp. 6-11. DOI: https://doi.org/10.1109/RAWSN.2015.7173270
  35. Pardey PG, Beddow JM, Hurley TM, Beatty TKM, Eidman VR, 2014. A bounds analysis of world food futures: global agriculture through to 2050. Aus J Agri Res Econ 58:571-589. DOI: https://doi.org/10.1111/1467-8489.12072
  36. Runge J, Saloux E, 2023. A comparison of prediction and forecasting artificial intelligence models to estimate the future energy demand in a district heating system. Energy 269:126661. DOI: https://doi.org/10.1016/j.energy.2023.126661
  37. Runge J, Zmeureanu R, 2019. Forecasting energy use in buildings using artificial neural networks: a review. Energies 12:3254. DOI: https://doi.org/10.3390/en12173254
  38. Serale G, Fiorentini M, Capozzoli A, Bernardini D, Bemporad A, 2018. Model predictive control (MPC) for enhancing building and HVAC system energy efficiency: problem formulation, applications and opportunities. Energies 11:631. DOI: https://doi.org/10.3390/en11030631
  39. Smith SL, Kindermans PJ, Ying C, Le QV, 2017. Don’t decay the learning rate, increase the batch size. arXiv:1711.00489.
  40. Song J, Zhang L, Xue G, Ma Y, Gao S, Jiang Q, 2021. Predicting hourly heating load in a district heating system based on a hybrid CNN-LSTM model. Energy Build 243:110998. DOI: https://doi.org/10.1016/j.enbuild.2021.110998
  41. Venkateswaran D, Cho Y, 2024. Efficient solar power generation forecasting for greenhouses: a hybrid deep learning approach. Alex Eng J 91:222-236. DOI: https://doi.org/10.1016/j.aej.2024.02.004
  42. Wang H, Asefa T, Duncan J, 2024. Event-based evaluation of operational ENSO forecasting models in 2002–2020: implications for seasonal water resources management. J Hydrol 636:131295. DOI: https://doi.org/10.1016/j.jhydrol.2024.131295
  43. Wang Y, Duan X, Zou R, Zhang F, Li Y, Hu Q, 2023. A novel data-driven deep learning approach for wind turbine power curve modeling. Energy 270:126908. DOI: https://doi.org/10.1016/j.energy.2023.126908
  44. Wang T, Mehdi QH, Gough NE, Griffiths IJ, 1997. A hybrid intelligent controller based on “consultation of doctors”. IFAC Proc 30:457-462. DOI: https://doi.org/10.1016/S1474-6670(17)43307-6
  45. Wei X, Liu Y, Gao S, Wang X, Yue H, 2019. An RNN-based delay-guaranteed monitoring framework in underwater wireless sensor networks. IEEE Access 7:25959-25971. DOI: https://doi.org/10.1109/ACCESS.2019.2899916
  46. Wu Y, Gao Y, Wang C, Chen Q, Ming T, 2023. The energy saving performance of the thermal diode composite wall in different climate regions. Renew Energy 219:119360. DOI: https://doi.org/10.1016/j.renene.2023.119360
  47. Yao Y, Shekhar DK, 2021. State of the art review on model predictive control (MPC) in heating ventilation and air-conditioning (HVAC) field. Build Environ 200:107952. DOI: https://doi.org/10.1016/j.buildenv.2021.107952
  48. Zhang Y, Meng F, Wang R, Zhu W, Zeng XJ, 2018. A stochastic MPC based approach to integrated energy management in microgrids. Sustain Cities Soc 41:349-362. DOI: https://doi.org/10.1016/j.scs.2018.05.044

Metrics

Citations and online attention from Dimensions, Altmetric and PlumX.